Analysis of the algorithm: From kernels to backup genes.

Kernelization section

The algorithm transformed the semantic similarity matrix to make it compatible with a kernel. Once this was done for each network and kernel type, it was integrated by kernel type. Below there is a general analysis of the properties of each matrix in the different phases of the process.

Matrix properties

Table 1. Similarity matrixes

Net Matrix_Dimensions Matrix_Elements Matrix_Elements_Non_Zero
gene2biological_process 100x100 10000 9898
gene2molecular_function 100x100 10000 9898
gene2cellular_sublocation 100x100 10000 9898
gene2phenotype 100x100 10000 9900

Table 2. Filtered similarity matrixes

Table 3. Uncombined kernel matrixes

Net Kernel Matrix_Dimensions Matrix_Elements Matrix_Elements_Non_Zero
gene2phenotype rf 100x100 10000 10000
gene2cellular_sublocation rf 100x100 10000 10000
gene2molecular_function rf 100x100 10000 10000
gene2biological_process rf 100x100 10000 10000
gene2phenotype ct 100x100 10000 10000
gene2cellular_sublocation ct 100x100 10000 10000
gene2molecular_function ct 100x100 10000 10000
gene2biological_process ct 100x100 10000 10000

Table 4. Integrated kernel matrixes

Integration Kernel Matrix_Dimensions Matrix_Elements Matrix_Elements_Non_Zero
mean rf 200x200 40000 20000
mean ct 200x200 40000 20000

Weight values

Comparing each layer

Comparing types of kernel

Comparing integrations and kernel types